Global Monsoon, El Nino, and Their Interannual Linkage Simulated by MIROC5 and the CMIP3 CGCMs

This study evaluates the capability of coupled global climate models (CGCMs) in simulating the prime examples of the forced response (global monsoon) and internal feedback process (El Nino). Emphases are also placed on the fidelity of the year-to-year variability of global monsoon precipitation that is coordinated by the interannual sea surface temperature (SST) fluctuation over the tropics. The latest version of the Model for Interdisciplinary Research on Climate 5 (MIROC5) with advanced physical schemes is compared with the two previous versions (MIROC3.2, high- and medium-resolution versions) and with the 20 CGCMs participating in the third phase of the Coupled Model Intercomparison Project (CMIP3). The climatological annual mean and cycles of precipitation and 850-hPa winds, the key components to demarcate the global monsoon domain, are reproduced better in MIROC5 than in MIROC3 versions. As a consequence, the former considerably outperforms the latter and is generally superior to the CMIP3 CGCMs in replicating the intensity and domain of global monsoon precipitation and circulations. These results highlight the importance of the improved physical parameterization in a model. Analyses of the monthly Nino-3 index suggest that the amplitude and periodicity of El Nino are simulated better in MIROC5 than in the MIROC3 versions. Yet the reality of nonlinear ENSO dynamics measured indirectly by the SST asymmetricity over the equatorial Pacific is unsatisfactory in the MIROC family as well as in the majority of the CMIP3 models. The maximum covariance analysis shows that a significant fraction of the interannual global monsoon rainfall variability is in concert with El Nino. The multimodel results reveal that such coupling is robust across the current CGCMs. More importantly, the fidelity of the global monsoon precipitation significantly relies on the realism of tropical SST. Comparison among the MIROC models suggests that improved El Nino is likely attributable to the more realistic Bjerknes feedback loop, which results from the intensified convective activity over the equatorial central Pacific Ocean.