Seasonal forecasting of east african rains

In East Africa (EA), rainfall variability significantly impacts socioeconomic and environmental conditions, making accurate seasonal predictions essential. Rainfed agriculture, vital for livelihoods and food security, is highly vulnerable to erratic rainfall, leading to lower yields and financial hardship. Global teleconnections like El Niño–Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD) strongly influence EA’s interannual rainfall variability, though their independent roles are not fully understood. We evaluated EA short rain predictability using C3S model ensembles, assessing 1- to 5-month lead forecasts initialized in September (1993–2016). Most models show skill in predicting OND precipitation anomalies but exhibit low skill in northern and western regions. Along Somalia’s coast and the western Indian Ocean, skill persists into late winter (DJF), likely due to SST anomaly persistence. Models outperform persistence forecasts during mature ENSO/IOD phases. The Dipole Mode Index predicts rainfall anomaly signs, confirming that broader-scale IOD variability associated with changes in the Walker Circulation, not local SST changes, drives EA rainfall. We also assessed long-rain predictability using C3S models initialized in February, evaluating at lead times from MAM to MJJ. Long rain variability is associated with ENSO, with models performing better during active ENSO phases than IOD-dominated periods. Consequently, the C3S seasonal prediction system demonstrates greater skill in reproducing the long rains during active ENSO phases compared to periods dominated by IOD variability. Using Community Earth System models (CESM) experiments, we examined ENSO and IOD’s independent roles in EA short rain variability. Partial correlation and composite analyses highlight IOD’s dominant influence, with warm (cool) SST anomalies linked to above (below) normal OND rainfall. ENSO’s direct impact is weaker and IOD-dependent. CESM_noENSO and CESM_noIOD experiments confirm IOD’s critical role, which persists even without ENSO variability, underscoring its independent influence on EA short rain variability.