Simulation and extended range prediction of monsoon intraseasonal oscillations in NCEP CFS/GFS version 2 framework

Abstract

The present study investigates the role of ocean– atmosphere coupling in improving the simulation and extended range prediction skill of the monsoon intraseasonal oscillations (MISOs) using the NCEP CFS (version 2) vis-a-vis its atmospheric component GFS (version 2) forced with bias-corrected sea-surface temperature (SST) derived from CFS. Though the CFS free-run analysis shows dry bias over Indian land as compared to GFS, the interactive air–sea coupling in CFS has considerably improved the simulation of large-scale dynamical fields, SST–rainfall relationship, and the northward propagation of the MISOs with respect to GFS. However, the improvement of MISO simulation in CFS over GFS has not necessarily guaranteed the improvement of real-time extended range prediction during 2011 and 2012. CFS shows better skill over GFS (forced with bias-corrected CFS derived SST) at pentad lead 4. The phases of MISOs are better predicted in GFS and the amplitude prediction skill is marginally improved in CFS. The present study also advocates the need of probabilistic category (active, normal or break) forecast at extended range.