Interannual Variability of the Asian-Australian Monsoon and ENSO Simulated by an Ocean－Atmosphere Coupled Model

This work assesses the performance of the new generation of LASG/IAP coupled climate system model named FGOALS_s in simulating the relationship between the Asian-Australian monsoon (AAM) and ENSO (El Niño-Southern Oscillation). Compared with the observations, FGOALS_s can simulate main characteristics of climatology of AAM. The amplitude of ENSO simulated by FGOALS_s is about 70% of that in observations. The FGOALS_s model reasonably reproduces the irregularity of ENSO period. Main features of ENSO are reasonably reproduced in FGOALS_s model, i.e., when SST in the Niño3 region is higher than normal, the sea surface pressure is higher than normal in the warm pool, but lower in the eastern Pacific. Phase locking of ENSO reproduced by FGOALS_s is in boreal spring or summer. This is the main deficiency of the model. The amplitude of interannual variability of AAM is larger than that in the observation. FGOALS_s fails in simulating the negative correlation between AAM indexes and contemporary Niño3 index, which may be caused by wrong phase locking of ENSO. When El Niño events reach peaks in boreal spring or summer, anomalous subsidence of Walker circulation moves eastward. An anomalous anticyclone stimulated by the anomalous subsidence is located to the east of that in the observation. As a result, the negative correlation between Indian monsoon rainfall and ENSO is not significant. During boreal winter, SST anomalies over the tropical eastern Pacific are weaker than those in observations, so AAM response is also weak. On the other hand, the divergent center, which is located over the Australian monsoon region in the observation, shifts westward in the model, because SST anomalies in the equatorial central and eastern Pacific extend westward, resulting in an insignificant negative correlation between the Australian monsoon rainfall and ENSO in the model. This analysis has provided a guide for the future improvement of the coupled model.