The simulation of subseasonal and interannual variability of the Indian monsoon in the latest coupled general circulation model, the Climate Forecast System (CFS), developed by the National Centers for Environmental Prediction is examined. The daily mean values of precipitation, outgoing longwave radiation (OLR), winds and sea surface temperature (SST) from a long coupled run are analyzed in the present study.

The space-time structures of the dominant modes of subseasonal oscillations and seasonally persisting patterns of rainfall, OLR and winds are isolated using multi-channel singular spectrum analysis (MSSA). The relative contribution of the different MSSA eigenmodes toward the seasonal mean monsoon is examined. The model's results are compared with observed monsoon variability that consists of 45-day and 20-day oscillations and persisting patterns related to El Niņo and Southern Oscillation and the Indian Ocean dipole patterns. From observations, it has been found that the seasonally persisting daily patterns of rainfall and winds are related to SST variations and provide hope for prediction of seasonal mean monsoon. The potential for such seasonal predictability of the CFS due to the slow variability of the boundary conditions is assessed by investigating the relation between the persisting modes and the daily SST generated by the model.