Abstract
In this study, a new methodology is developed to improve the climate simulation of state-of-the-art coupled global climate models (GCMs), by a postprocessing based on the intermodel diversity. Based on the close connection between the interannual variability and climatological states, the distinctive relation between the intermodel diversity of the interannual variability and that of the basic state is found. Based on this relation, the simulated interannual variabilities can be improved, by correcting their climatological bias. To test this methodology, the dominant intermodel difference in precipitation responses during El Niño–Southern Oscillation (ENSO) is investigated, and its relationship with climatological state. It is found that the dominant intermodel diversity of the ENSO precipitation in phase 5 of the Coupled Model Intercomparison Project (CMIP5) is associated with the zonal shift of the positive precipitation center during El Niño. This dominant intermodel difference is significantly correlated with the basic states. The models with wetter (dryer) climatology than the climatology of the multimodel ensemble (MME) over the central Pacific tend to shift positive ENSO precipitation anomalies to the east (west). Based on the model’s systematic errors in atmospheric ENSO response and bias, the models with better climatological state tend to simulate more realistic atmospheric ENSO responses.
Therefore, the statistical method to correct the ENSO response mostly improves the ENSO response. After the statistical correction, simulating quality of the MME ENSO precipitation is distinctively improved. These results provide a possibility that the present methodology can be also applied to improving climate projection and seasonal climate prediction.
