Assessing precipitation seasonal forecasts in Central Africa using North American Multimodel Ensemble (NMME)

Abstract

This study examines the seasonal forecast of the North American Multi-Model Ensemble (NMME) over Central Africa (CA), which encompasses a region of the world where the economies of the countries are heavily dependent on agriculture and livestock. The forecast evaluation is done over the 4 seasons: December to February (DJF), March to May (MAM), June to August (JJA), September to November (SON) between 0 to 5 months lead time before the beginning of each season corresponding to many regional climate perspectives. Deterministic and categorical approaches focusing on the rainfall variable were used to assess NMME models over CA. The observed and predicted rainfall have been divided into three categories: below normal, normal and above normal. The results show that for 0 to 2 months lead time, the NMME Multi-model Ensemble Mean (MME) reproduces rainfall peaks (9 to 10 mm / days) at the level of Atlantic coast and in East of Democratic Republic of Congo in MAM and SON. Always between 0 to 2 months lead time, values of correlation coefficients (R) of all seasons are greater than 0.72. For 3 to 5 months lead time, low values of R are observed. It follows that probabilities of detection (POD) are greater than 50\% for all different normal seasons and less than 45\% for below and above normal seasons. On the other hand, high false alarm (FAR) values and low Critical Success Index (CSI) values are observed for both below and above normal seasons. The NMME model emerges as an interesting tool capable of providing important seasonal characteristics up to several months (5 months) before the start of the season in the CA, which will allow proper consideration of meteorological phenomena.