Author(s)

Edmund Mutayoba^1,2*, Japhet J. Kashaigili³

¹Department of Built Environmental Engineering, Mbeya University of Science and Technology, Mbeya, Tanzania.
²Department of Engineering Sciences and Technology, Sokoine University of Agriculture, Morogoro, Tanzania.
³Department of Forest Resources Assessment and Management, Sokoine University of Agriculture, Morogoro, Tanzania.

This study aims to evaluate the performance of the individual Regional Climate Models (RCMs) used in Coordinated Regional Climate Downscaling Experiment (CORDEX) and the ensemble average of the four RCMs to feign the characteristics of the rainfall pattern for the Mbarali River catchment in Rufiji Basin for the period of 1979 to 2005. Statistical analysis for model performance such as Root mean square error, Mean error, Pearson correlation coefficient, Mean, Median, standard deviation and trend analysis are used. In addition to the statistical measure of model performance, the models are tested on their ability to capture the observed annual cycles and interannual variability of rainfall. Results indicated that the RCMs from the CORDEX indicated a better performance to reproduce the rainfall characteristics over Mbarali River catchment in Rufiji Basin. They reproduced fairly the Era Interim annual cycle and inter-annual variability of rainfall. The ensemble average performed better than individual models in representing rainfall over Mbarali River catchment in Rufiji Basin. These suggest that rainfall simulation from the ensemble average will be used for the assessment of the hydrological impact studies over Mbarali River catchment in Rufiji Basin.

Climate Change, CORDEX, Regional Climate Models (RCMs), Ensemble Average

Mutayoba, E. and Kashaigili, J. (2017) Evaluation for the Performance of the CORDEX Regional Climate Models in Simulating Rainfall Characteristics over Mbarali River Catchment in the Rufiji Basin, Tanzania. Journal of Geoscience and Environment Protection, 5, 139-151. doi: 10.4236/gep.2017.54011.