Abstract
Global Circulation Models (GCMs) are widely utilized in climate projection. Nonetheless, the resolution of GCM products cannot satisfy local studies because the spatial resolution of GCMs is too coarse to represent regional climate variations at scales required making it difficult to provide precise information at a regional scale. Downscaling was developed to bridge the deficiency of GCM’s in climate studies. Therefore, this study was aimed at validating the usefulness of statistical downscaling model for rainfall prediction under different emission scenarios over the south-south region of Nigeria. The expost-factor research design was adopted for the study. The quadrat sampling technique was adopted by stratifying the area into 2°x 2°latitude and longitude intersections and each weather station that fall within the grids calibrated (Asaba, Warri, Uyo and Port Harcourt) were selected. Rainfall data for the selected meteorological stations covering 1985-2015 were acquired from the archives of Nigerian Meteorological Agency (NiMet) while large scale predictors were assessed from the archives of National Centre for Environmental Prediction (NCEP). Based on the regression analysis performed, shum, rhum, r850, r500, p5_u, p_u, & p5th were selected are as the principal large scale predictors of rainfall in the area whereas p5th (0.74), r850 (0.78) were identified as super predictors in Asaba, while r850 (0.74), (0.72) and (0.77) were noted for Warri, Uyo and Port Harcourt respectively. The model model’s performance in terms of predictions of the predictand was assessed on seasonal scale of four compartmentalized seasons, December, January, February (DJF), March, April, May (MAM), June, July, August (JJA) and September, October, November (SON).The R and RMSE values which forms the basis of assessment of the model ranges between, R (0.64-0.91) and RMSE (0.11-0.43). The model’s performance was better in Asaba at DJF, Warri in JJA while Port Harcourt and Uyo perform better in SON. Thus, it was concluded that the model has the capability to accurately predict rainfall in the region on a seasonal timescale. Based on the findings of the study, development of a local climate management system in preparedness for climate change, efforts to maintain strategies to climate change were recommended.