Evaluation of the performance of IAP-AGCM4.1 in simulating the climate of West Africa

Abstract

Simulated sea surface temperature (SSTs) constitutes error source in tropical climate, simulated by coupled general circulation models (GCMs). Therefore, atmospheric GCMs that use observed SSTs are expected to simulate tropical climate with reduced bias. This paper evaluates the performance of the atmosphere only GCM of the Institute of Atmospheric Physics, Chinese Academy of Sciences (IAP-AGCM4.1) in simulating the climate of West Africa. Monthly simulated climatology of precipitation and temperature in five precipitation regions are compared with observed CRU climatology. Inter-annual variability in simulated precipitation and temperature are also compared with observation. In addition, the 10th, 50th and 90th percentiles of simulated precipitation and temperature are compared with the observed precipitation and temperature using normalised deviation from mean and normalised root-mean-square difference. Finally, observed and simulated links between multivariate ENSO index (MEI) and precipitation (temperature) are compared. IAP-AGCM4.1 simulates the mean climatology well over Western part of West Africa (WWA), Central Guinea Coast (GC), and Eastern GC. At the Eastern Sahel, precipitation is over estimated at the 10th and 50th percentile. Generally, there is overestimation at the 10th percentile and underestimation at the 90th percentile. The model also simulates the observed annual variability reasonably well. As expected positive MEI leads to reduction in precipitation in substantial part of West Africa while negative MEI leads to wetness. Furthermore, temperature increases with positive MEI while it decreases with negative MEI over West Africa. However, the model reproduces the dynamics of the influence of MEI on the climate of West Africa with weaker signal.