Otieno George*, Franklin Opijah*, Joseph Mutemi*, and Laban Ogallo*
*University of Nairobi, Department of Meteorology, Kenya
(Received 16 Mar 2018, received in revised form 18 September 2018, Accepted 19 September 2018)
The study investigated the ability of European Centre Global Model (GME), to reproduce the seasonal mean rainfall, inter-annual rainfall variability and annual rainfall cycles over the Greater Horn of Africa (GHA). The seasons considered were March - May (MAM) and October -December (OND) rainfall seasons. The analyses of the model output were done over the Northern East Africa (NEA), Southern East Africa (SEA), Central East Africa (CEA) and Eastern East Africa (EEA) sub-domains. The gridded observation datasets for model validation were from Climate Research Unit (CRU) and Global Precipitation Climatology Center (GPCC). The bias correction techniques as well as other spatial and temporal analyses were applied as metrics of evaluation. The mean rainfall simulations had closest agreement with the observations during the OND than MAM season. The GME model succeeded in reproducing the main rainfall peaks during the OND rainfall season but failed to capture the double peak rainfall over the equatorial sector of the GHA region during MAM season. The model reproduced most of the drier/wetter rainfall years associated with the El Niño Southern Oscillation (ENSO) events. The model exhibited large uncertainties in getting the anomalies associated with the large global forcing like ENSO events over the CEA and EEA sub-domains. Based upon the first law of thermodynamics, the GME model provides improved and better climate information for disaster risk reduction over the GHA region. This study is therefore a benchmark for future consideration of GME model for customization to provide large-scale information within a regional climate model for downscaling purposes.